At 7:51 AM 3/3/5, Robin van Spaandonk wrote: >In reply to Horace Heffner's message of Wed, 02 Mar 2005 10:52:52 >-0900: >Hi, >[snip] >>It does say, "The kinetics of the process is unclear." though. The bubbles >>of methane form on grain boundaries, and this requires diffusion of the >>carbon. >[snip] >Perhaps, it is the methane which migrates after it is formed, not >the carbon before the methane is formed.
The methane molecule would be a bit large to fit through the lattice. Here is an interesting table of atomic radii in angstroms: Atom Covalent bonded H 0.79 0.32 C 0.91 0.77 Si 1.46 1.11 The covalent radius of carbon, 0.77 angstroms, is slightly less that the stand-alone atomic radius of hydrogen, 0.79 angstroms. Adsorbed hydrogen generally does not have room to fit in a lattice site, though it is close, and this is evidenced by the degree of metal swelling as 1-1 hydrogen loading is approached. The ionically bonded electron associated with the adsorbed hydrogen nucleus forms a "partial orbital" which maintains pressure on the lattice, and vice versa. By partial orbital it is meant a fairly large probability of finding the paired electron in an orbital and the complimentary probability of finding the paired electron in a conduction band. The adsorbed hydrogen volume is thus slightly reduced from that of an atom with a 0.79 angstrom radius. Now for some conjectures. It is notable that the covalent carbon radius is slightly less than the hydrogen atomic radius. This means that carbon should be able to diffuse through a lattice about as easily as hydrogen, provided adjacent metal atoms can easily exchange covalent bonds with the carbon so as to allow it to advance when a pressure gradient is present, as in the close vicinity to a crack in the metal. It seems like some ways to reduce hydrogen embrittlement might be to avoid carbon steels, or alloys having metals the form hydrides, like Ni or Pd. Also, quenching in liquids conatining hydrogen or carbon may not be so good. It may be that LN would be a good quenching agent, but since it has an atomic radius of only 0.75 angstroms, it would be suspect for forming ammonia bubbles in a hydrogen loaded lattice similar in properties to methane bubbles. Silicon can form silane gas, similar to methane. However, silicon's atomic and covalent radii prevent it from diffusing. Silicon steels thus migth be a good choice for avoiding fast embrittlement if other bad things are not present in the alloy. If hydrogen can get into any material, however, it just seems like some lattice damage is likely to result. One way to keep hydrogen out is to impose a barrier. For room temperature applications a copper coating might do the trick. Copper can readily adsorb H at 600 deg. C though, and if quickly cooled lots of tiny spherical bubbles of hydrogen are formed in the copper. This might offer a useful way to prepare CF electrodes for bombardment, by loading at high temperature and cooling prior to particle or x-ray bombardment or other fusion triggering means. It also raises the question of what additional or unsuspected metals might be CF active at ceramic oven temperatures. New possibilites for annealing and reloading are also provided by use of high temperatures. Regards, Horace Heffner
